Existing projector systems typically project a dynamic and/or static image onto a surface and focus the image based autofocus algorithms that monitor one specific point or region of the image, e.g. a dead center of the projected image. This type of autofocus works sufficiently because the projection source is not expected to continuously change relative to the projection surface. However, in the case of a highly mobile projection surface or source, performing an autofocus without adversely impacting user experience becomes more difficult.
With the projector or surface reflecting the projected image in motion, the projected image may become distorted in any of three dimensions, and also may be projected on a surface at various angles. Moreover, the projected image typically has regions of interest to the viewer that might not be within the region of the image that the static autofocus is monitoring. Consequently, the use of an autofocus on a static region within the projected image may not adequately compensate for the motion of either the projector or surface, and does not necessarily keep the region of most likely interest to the viewer of the image in focus.